Evidence is accumulating that supports the concept that environmental carcinogenesis in rodents and humans is a time-dependent multi- genic process. Also, the importance of the inheritance of specific genetic susceptibility as a critical determinant of the disease process is being recognized. The primary assay for the identification of potential human carcinogens is the two year rodent bioassay, where inbred mice and rats are used as surrogates. We have proposed to supplement the two year cancer bioassays with rapid identification bioassays using transgenic mouse models. Two different groups of NTP chemicals are currently under evaluation in the p53-deficient (+/-) and Tg.AC mice. The first group chosen by NIEHS/NTP scientists contain both human carcinogens as well as rodent mutagenic and nonmutagenic carcinogens and noncarcinogens. The results from this group should provide information on the ability of these mouse lines to identify as well as discriminate between carcinogens of differing potency. A second group of chemicals represents another approach for validation of these models: the prospective testing by the rapid bioassays of chemicals that are currently being tested in the NTP cancer two year bioassay. Prospective analysis provides the opportunity to remove the potential bias of selection of chemicals for validation. A high correspondence between this short-term cancer bioassay and the long-term bioassay (of both groups) would provide assurance that mutagenic and nonmutagenic carcinogens and noncarcinogens may be identified using these mouse models. However, many critical questions remain to be answered in order to promote a scientific consensus on the use of these transgenic lines for identification of potential human carcinogens.